All player stats - data from the KSoD website

General topics, questions, suggestions, bugs, or anything Star Wars Commander related.
DrPain
Commodore
Posts: 200

Re: All player stats - data from the KSoD website

Post#71 » Sun Dec 03, 2017 9:06 am

YourFather wrote:
DarkRebel wrote:
YourFather wrote:
It's interesting because just about every rebel on here is saying imperial gr snipers are the reason imps are dominating conflicts. If that were true, then we would see it in the data. We don't see it, we haven't seen it and we won't see it because the claim is completely false.

Sorry OP, I know you want armory discussions elsewhere so I'll respond to the rest of DR's post;

I'm not sure if average medals per win is even a good indication of balance. Imperials are primarily hitting dev bases during conflicts. We are pretty much guaranteed 3 stars per attack and ~15 medals. Rebels face tough bases and occasionally faceplant. Sure, tough bases can offer high medals but this is more of an exception than a standard.

It's hard to conclude medals are a state of balance in the game when the factions experience completely different offensive environments.

You must have presumed when I said balance, that it implied unit balance. Bad assumption. When I said balance, it meant in an overall sense.
Of course match making matters in game experience, in how easier it was for IMP to play and to gain medal. You said it yourself.
As game stats doesn’t track losses, of all the stats available to us, medals per win represents the best overall status as it is the RESULT of PVP, and everything like match making, losses (results in negative medals), base difficulty level, dev bases, armory level, are all partially factored in one way or another.
(last post on this, to respect the OP wish of not debate on game balance)



I don't think my post left any room assumptions. Bad assumption on your part :)

All I've said is that the data clearly and unequivocally show that rebels are doing much better than imperials on defense. They are also dominating squad wars (despite all the imbalanced matchmaking threads they write). You're not arguing with me, you're arguing with the numbers and that's just silly.


And those def wins are out of how many defenses? If you can't answer that question, then you can't really show anything.

I know that individual players have been able to find out the number of def and attacking losses for their acct. But those numbers aren't on the ksod site.


User avatar
YourFather
Commodore
Posts: 162

Re: All player stats - data from the KSoD website

Post#72 » Sun Dec 03, 2017 12:59 pm

ObiWanKenobi2016 wrote:
YourFather wrote:It's interesting because just about every rebel on here is saying imperial gr snipers are the reason imps are dominating conflicts. If that were true, then we would see it in the data. We don't see it, we haven't seen it and we won't see it because the claim is completely false.


No. These stats do not assist with the effectiveness or otherwise of Imp / Reb GR, or indeed any other update.

The KSoD bot trawls the squads recording the number of attack wins and defence wins (among other data). These stats therefore reflect wins and losses (for some players), since the game started. Due to their longevity in the game these commanders' stats will have a significant distorting affect (perhaps especially so for the minority faction). The game has been going for more than 3 years now. Commanders that have kept with the game for all that time will have large numbers of wins and losses because of how the game used to be, not because of how it is now or how it was during "the OP Imp GR era".

Please note, I am describing the era where Imp GR was stronger as the 'OP Imp GR Era', simply because the devs decided to nerf it to it's current levels. Only the devs are able to access data that might show the actual impact on the game. They decided to nerf it, and buff reb GR. I don't know of anyone who was asking for Reb GR to be buffed at all, let alone in the way it was buffed. What Rebs were asking for was to have their own version of the sniper GR. The devs went a different way.

I think Dark Rebel's theory for the stat difference is right, it' more to do with the (much) higher number of Imp players. On a normalcy curve there are far more poor players than good ones. With the 3:2 faction ratio there will have been a lot more weak imps over the course of the game's life who have failed on attacks and had weaker bases when attacked by rebels. Over 3 years these numbers stack up. This explains the stat discrepancy on rebel attack / defence success much better. Others may have other views, of course.

The only real guide to the current game balance is to be found in conflict leaderboards, and even then, only where the prize is of equal value for both factions - eg during conflicts for jump/jet skins. A conflict where one faction has a great prize, and the other a rubbish one does not assist, eg Imp Astromech v Reb WED. Only conflict leaderboards demonstrate attack and defence success during a specific time frame in a way that we can see.


I agree you can't look at the lifetime of wins as a guide for the current state of the game. One would look at the CHANGES in wins between each update you have so diligently provided us. Sure, some people go inactive between the updates but any sort of statistical anomaly (say from OP GR) would appear in these changes.

Also, I have to point out the very large disparity in wins vs losses between imperials and rebels. You can say players in faction cause an imbalance all day but a statistical difference that large would likely need far more explaining.

Lastly, the conflict leaderboards are truly the absolute worst guide for the state of balance in the game. 14 out of 50 players in the last leaderboards were using macros, that's already more than a quarter of the leaderboard that shouldn't even be considered (but many forum rebels did factor them in ) for game balance.

From there you have to consider the unfair matchmaking. Imperials have learned to identify and hit dev bases all conflict while rebels are forced to hit the toughest player bases. Of course imperials are dominating conflicts (except on tatooine where we don't get the luxury of dev bases). I can level a max 10 dev base with 8 jets (no hero) for 640 points then almost immediately do it again when I get back to base.

You see, matchmaking gives imperials the same advantage over rebels that macros players have over us all; time vs point accumulation. This is far worse a problem than sniper gr ever was. How many points per attack do you reasonably look to gain? How long do you wait for your troops to cook between attacks? Still want to tell me sniper gr is the reason I'm getting more points than you?


User avatar
LosVega
Major General
Posts: 342

Re: All player stats - data from the KSoD website

Post#73 » Sun Dec 03, 2017 5:40 pm

DarkRebel wrote:I know that individual players have been able to find out the number of def and attacking losses for their acct. But those numbers aren't on the ksod site.

Solely out of curiosity, how? Data mining or is Disney actually responding to that kind of query?
Image
BATTALION FAMILY
WOLF, DRAGON, REDBAK, FALCON, GATOR
Duza's Elite - Empire
Looking for a great squad? Come by for a visit, make new friends and maul
some ass


DrPain
Commodore
Posts: 200

Re: All player stats - data from the KSoD website

Post#74 » Sun Dec 03, 2017 5:54 pm

LosVega wrote:Solely out of curiosity, how? Data mining or is Disney actually responding to that kind of query?

I'm not sure how, but I've known a few people who have gotten it somehow. I know that it's not through support since that person was able to get the information for a friend on his behalf. I'm intentionally leaving out the names to make sure that noone gets in trouble (although I doubt they would anyway).


wedge
Commander
Posts: 26

Re: All player stats - data from the KSoD website

Post#75 » Sun Dec 03, 2017 7:54 pm

I think a one number average is a terrible measure and no useful information can be found from any of the numbers there. It is an average it doesn't account for differences between number/percentage of hardcore players, play styles, time spent in game, how long they have played


legofanatikeren
Commodore
Posts: 136

Re: All player stats - data from the KSoD website

Post#76 » Mon Dec 04, 2017 12:13 am

We must agree to be inconclusive, then move forward. We can argue about how and what and why all year long.

It is interesting to see the development, no matter how it is interpreted, and what the "right" interpretation is. Nice inputs though.
legofanatikeren Was Here. Proud member and officer of The jedi-Family.
#TeamLeo #flæskesteg #DanishHumor
If I meet you - you better like LEGO! :D
... No, seriously - you better do.


User avatar
Burrito
Commodore
Posts: 137

Re: All player stats - data from the KSoD website

Post#77 » Tue Dec 05, 2017 6:12 am

I think they are just using the baseline of squads and extrapolating the active members/wins/everything else out of that. Limited information but a good use of the information available.


User avatar
ObiWanKenobi2016
Admiral
Posts: 629

Re: All player stats - data from the KSoD website

Post#78 » Tue Dec 05, 2017 1:15 pm

YourFather wrote:
I agree you can't look at the lifetime of wins as a guide for the current state of the game. One would look at the CHANGES in wins between each update you have so diligently provided us. Sure, some people go inactive between the updates but any sort of statistical anomaly (say from OP GR) would appear in these changes.


I think that would have been right if we were looking at data in the early days of the game, but we've started collecting it too late now. A 3:2 ration imbalance is significant in one day's worth of data. 3 years on, it's all distorted out of proportion. Just to make a statistic difference would require huge numbers of successful attacks / wins. You only have to think briefly about the number of attacks a player base of 260,000 might make in a day, especially during a conflict to realise how hard it is to shift those numbers at this time.

If you look at the average number of medals, you may find a better indication of the game's balance. Despite the apparent better attack stats, the medal score is quite similar. This suggests the rebel wins don't earn many medals for their wins, suggesting it's more 1 -2 star wins for the rebels, when the Imps are getting 2-3? However, as I say, we're too far down the line to be able to make much of these changes.

YourFather wrote:... Lastly, the conflict leaderboards are truly the absolute worst guide for the state of balance in the game. 14 out of 50 players in the last leaderboards were using macros, that's already more than a quarter of the leaderboard that shouldn't even be considered (but many forum rebels did factor them in ) for game balance.

We're essentially talking about one squad's unusual playing style. They do not accept they are using macros to game the system, and they are no longer posting the extraordinary numbers of attacks that they once did. Whether they are not devoting as much of their life to this game, or switching off the macro early we cannot know. But I agree they should be discounted from any assessment of game balance. Their unusual playing style does not assist with an assessment of how the game is for those playing conventionally.

YourFather wrote:From there you have to consider the unfair matchmaking. Imperials have learned to identify and hit dev bases all conflict while rebels are forced to hit the toughest player bases. Of course imperials are dominating conflicts (except on tatooine where we don't get the luxury of dev bases). I can level a max 10 dev base with 8 jets (no hero) for 640 points then almost immediately do it again when I get back to base.

You see, matchmaking gives imperials the same advantage over rebels that macros players have over us all; time vs point accumulation. This is far worse a problem than sniper gr ever was. How many points per attack do you reasonably look to gain? How long do you wait for your troops to cook between attacks?


These are all good points, although I must note that when rebels dominated Erkit a year or so ago and we also saw dev bases. 'Identifying' them was never hard. They all had Very low resources, (and in those days) no live dekas. These days it's still low resources, but no glowing barracks or factories.
HQ 10 - 15K medals - Kenobi's Command - L 50 Squad - Max perks - 3 Squad Wars a week

Stop by for Wars, perks and rep points. We're maxed out.

"You must do what you think is right, of course" (We're the good guys)


User avatar
YourFather
Commodore
Posts: 162

Re: All player stats - data from the KSoD website

Post#79 » Tue Dec 05, 2017 1:46 pm

ObiWanKenobi2016 wrote:
YourFather wrote:
I agree you can't look at the lifetime of wins as a guide for the current state of the game. One would look at the CHANGES in wins between each update you have so diligently provided us. Sure, some people go inactive between the updates but any sort of statistical anomaly (say from OP GR) would appear in these changes.


I think that would have been right if we were looking at data in the early days of the game, but we've started collecting it too late now. A 3:2 ration imbalance is significant in one day's worth of data. 3 years on, it's all distorted out of proportion. Just to make a statistic difference would require huge numbers of successful attacks / wins. You only have to think briefly about the number of attacks a player base of 260,000 might make in a day, especially during a conflict to realise how hard it is to shift those numbers at this time.

If you look at the average number of medals, you may find a better indication of the game's balance. Despite the apparent better attack stats, the medal score is quite similar. This suggests the rebel wins don't earn many medals for their wins, suggesting it's more 1 -2 star wins for the rebels, when the Imps are getting 2-3? However, as I say, we're too far down the line to be able to make much of these changes.

YourFather wrote:... Lastly, the conflict leaderboards are truly the absolute worst guide for the state of balance in the game. 14 out of 50 players in the last leaderboards were using macros, that's already more than a quarter of the leaderboard that shouldn't even be considered (but many forum rebels did factor them in ) for game balance.

We're essentially talking about one squad's unusual playing style. They do not accept they are using macros to game the system, and they are no longer posting the extraordinary numbers of attacks that they once did. Whether they are not devoting as much of their life to this game, or switching off the macro early we cannot know. But I agree they should be discounted from any assessment of game balance. Their unusual playing style does not assist with an assessment of how the game is for those playing conventionally.

YourFather wrote:From there you have to consider the unfair matchmaking. Imperials have learned to identify and hit dev bases all conflict while rebels are forced to hit the toughest player bases. Of course imperials are dominating conflicts (except on tatooine where we don't get the luxury of dev bases). I can level a max 10 dev base with 8 jets (no hero) for 640 points then almost immediately do it again when I get back to base.

You see, matchmaking gives imperials the same advantage over rebels that macros players have over us all; time vs point accumulation. This is far worse a problem than sniper gr ever was. How many points per attack do you reasonably look to gain? How long do you wait for your troops to cook between attacks?


These are all good points, although I must note that when rebels dominated Erkit a year or so ago and we also saw dev bases. 'Identifying' them was never hard. They all had Very low resources, (and in those days) no live dekas. These days it's still low resources, but no glowing barracks or factories.


Thanks for the thoughtful reply. I'm not sure I agree if there's too many data points that an anomaly wouldn't present itself but I can't disprove it either. I guess at the end of the day we're two guys on the internet speculating about what data show. I do agree though that average medals per player wouldn't be a good diagnostic of current game health if we are buried in data points.

Agreed. Outspoken macros players shouldn't be considered in game balance data. They also take up as much as (from past conflicts) 28% of the conflict leaderboard. Hence, the leaderboard as a diagnostic of balance, is already in jeopardy from that fact alone.

I'm glad you brought up the days of rebellion dominance on Er'kit as it proves my point exactly. Rebels completely dominated the kitty leaderboard those days. No wonder, due to their high population they were given easier matchmaking while imperial had to slog through the mud to get a star or 2 and the occasional 3. The effects of this are still felt when you see so many imperials with a low level kessen (something that is finally being remedied).

Eating dev bases these days gives an greater advantage than it ever did thanks to the armory. Attacking dev bases at lighting speed with partial loadouts and still gaining over 600 points for a 3 star? It's a no-brainer.


User avatar
acidrave
Captain
Posts: 60

Re: All player stats - data from the KSoD website

Post#80 » Tue Dec 05, 2017 11:44 pm

YourFather wrote:

I'm glad you brought up the days of rebellion dominance on Er'kit as it proves my point exactly. Rebels completely dominated the kitty leaderboard those days.



not true: with 12 hours to go, there are 5 in the top 50
REBEL LEVEL 9
PROUD MEMEBER OF JARJAR_CAMP
http://ksod.us/squad/3a980df2-878d-11e5-82e3-063228004ed8/members?hl=863e1f57-e6eb-11e6-90ae-06cad0004ecc


Return to “General Discussions”

Who is online

Users browsing this forum: aug13, CommonCrawl [Bot], sgu97cab, SpaceDandy, TSQL and 34 guests